As 2025 draws to a close, the tech sector has undergone significant transformations, driving tech stocks to become the market's dominant theme throughout the year. However, year-end volatility has intensified in AI, semiconductors, and new energy sectors. With renewed debates about an AI bubble, the investment roadmap for the coming year has become a focal point for investors.
Amid the uncertainty, Great Wall Fund hosted its 2026 Investment Strategy Conference. Han Lin, manager of the Great Wall Digital Economy Fund and a seasoned tech investor, addressed hot topics such as the "AI bubble debate" and "tech investment strategies for 2026."
Han Lin noted that discussions about an AI bubble have gone through three cycles since 2023, with each wave of concern gradually dissipating as the industry advanced. Initial doubts in 2023 about the sustainability of AI investments were dispelled by sustained capital expenditures and cloud revenue growth from major overseas CSP providers. Later concerns about computing power deflation from late 2024 to mid-2025 were alleviated as North American markets continued their "brute-force" approach to model training.
Recent worries about cyclical investments have also eased, as leading CSP players developed high-quality models using in-house ASIC chips, establishing independent ecosystems. Based on these trends, Han Lin concluded that "AI remains in the relatively early stages of a tech wave, with little evidence of a bubble."
Beyond the bubble debate, a more critical question emerges: What phase is the AI industry in today? This directly shapes the foundational logic for investment. Comparing AI to historical tech cycles, Han Lin emphasized its early-stage characteristics. While both AI and the 2000 internet boom belong to the global TMT wave, AI demonstrates stronger fundamentals and aligns more closely with early-growth traits.
"Currently, the market is still in a phase of heavy infrastructure development, with monetization and commercialization still being explored. This suggests we're in the early-growth stage," he said. He further stressed that the core investment thesis for AI remains its untapped growth potential.
For retail investors assessing sector cycles and inflection points, Han Lin proposed a three-dimensional framework: 1. Supply-demand dynamics—strong demand and tight supply typically signal an upcycle. 2. Financial metrics—improving gross margins, net margins, and ROE indicate cyclical strength. 3. Earnings trajectory—companies often progress from underperforming to consistently exceeding expectations. In AI specifically, tracking GPU manufacturers' production schedules and CSP providers' capex guidance can validate cycle conditions.
Looking ahead to 2026, how should opportunities and risks across the AI value chain be positioned? Han Lin identified upstream computing infrastructure as the highest-conviction play: "We're in the midst of a computing infrastructure arms race." He highlighted shortages across GPUs, computing, connectivity, storage, and power equipment as creating abundant investment opportunities. Key challenges include supply chain bottlenecks and geopolitical uncertainties.
For midstream model/platform players, next year's battleground lies in ecosystem positioning. Model advancements could empower platform companies to gain market share, while major CSPs—serving as both model owners and cloud providers—may monetize through integrated SaaS/PaaS offerings. However, cost pressures (from compute, power, and R&D) and rapid technological obsolescence pose significant risks.
Downstream applications present a fragmented landscape, with Han Lin favoring AI+SaaS for enterprise use: "Efficiency gains in productivity tools strengthen B2B payment willingness, enhancing commercial viability." The main hurdle here is commercialization—weak model-application alignment or prolonged adoption cycles may dampen monetization.
Disclaimer: This communication contains information from sources believed to be reliable, but accuracy/completeness isn't guaranteed. Views expressed may change without notice. This isn't investment advice. Neither Great Wall Fund nor its affiliates assume liability for losses arising from its use. Unauthorized reproduction/distribution is prohibited. Investing involves risks.
Comments